DC generator issues after rewind

[eeprom]

Hello,
I have a DC compound wound self-excited generator which was recently rewound with a new armature and new commutator. The shunt and series fields were not changed.

The shunt field is wired in parallel to the series field through a rheostat. At normal operation, the rheostat used to operate in one particular position - I don't know the exact ohms, but there is a marking on the panel. But now in order to achieve the same output voltage, the rheostat has to be set much lower. So the shunt field needs more current than it used to in order that the gen can produce the same output voltage.

The brushes are well seated. The machine has been running for two weeks. There is no arcing that I have heard of. I don't have the opportunity to shut the machine down.

The only things I can think of is that the rotor was wound improperly, as in the windings are not all the correct polarity, or the brush angle is off because of the new commutator.

If the new commutator was smaller than the old commutator, resulting in the brushes touching one too many com bars, what would that do? Could that cause this problem? If so, are there any patterns on the com bars which would reveal/indicate this problem?

thanks,
EE

 

[Motorwinder]

Can you get winding data from the vender that wound it? It might have been a duplex winding, and they rewound it to a simplex.

Also,the interpole polarity might be set wrong. But that would cause sparking.

If you can post all the nameplate data, I'll check and see if I've wound one before.

 

[eeprom]

Unfortunately I don't have the nameplate data. This is a very old machine. It is used to excite a rotor rated at 150A and 253V. It is compound wound. The winding is a wave winding. It appears there are two coils per slot. There are four sets of brushes. The brushes are aligned to the neutral position as marked on the brush holder.

I realize this is not a lot of information, but that's all I have right now.

EE

 

[rasevskii]

The brushes may not be in the correct neutral position, that is, the neutral position for the new arm. winding is slightly different from the old winding.

A check can be made (unit shutdown) by applying an AC voltage across thr shunt field (say 110V) and shifting the brushholder assembly until zero millivolts AC is read across the brushes. External circuits disconnected. This will give the new neutral point, which can be marked on the brushholder assembly.

BTW you said the shunt field is parallel to the series field. That is wrong. Check the connections again.

It can be that the new winding has a different turn count from the old also.

rasevskii

 

[Motorwinder]

If the rheostat is set at a lower setting, wouldn't that be more resistance? Seems that would be lower voltage to the shunt field.

If the armature had some shorts before rewinding, Having a new winding, it would increase the output for a given excitation.

The commutator diameter wouldn't have an effect on the output, or the brush angle. As long as the brushes are fully seated, and spanning all the bars it should.

You might check the neutral by exciting the shunts with AC current, and measuring across your brush holder for the lowest voltage.

 

[eeprom]

Correction...the shunt field is in series with a rheostat. Those two devices together are in parallel to armature (A1 A2). A1 is wired to the brushes at -45 and 135 degrees, brushes at 45 and 225 degrees are wired to the series field. The other side of the series field comes out at A2.

The AC test sounds like a very good idea, despite the fact that I'd have to shut down the unit. Would a small variation, like 1 or 2 degrees, cause this much difference in operation?

The rheostat is in series with the shunt field. The smaller the resistance of the rheostat, the more current goes to the shunt field.

 

[Motorwinder]

Two things could've thrown the neutral off.
Replacing the commutator, and rewinding.

Comms are aligned to either a bar or a mica segment. If the comm isn't keyed it's easy to get it off. That would throw it off by a half a bar.

If the winder didn't line out his data right. The winding could be off by a bar. Skewed slot generators are easy make a mistake on.

 

[eeprom]

If the neutral was off, wouldn't there be sparking?

 

[Motorwinder]

Normally yes. On DC motors at least. Not sure about generators. It could have a strong compensating circuit? Engineer might chime in.

 

[eeprom]

The new commutator is 7.375". The previous commutator was 6.875". How much of a difference might that make?

 

[Motorwinder]

A lot longer comm life. It's not going to effect performance as long as you reset your neutral.

 

[eeprom]

What if the shunt field was wired backwards?

 

[rasevskii]

If the shunt field was wired backwards, the generator would not self-excite as the remanace field in the iron would be the opposite polarity. Or if the unit would self-excite it would have the wrong output polarity.

rasevskii

 

[eeprom]

Follow up on this thread....

I did some more investigating and found the smoking gun. The shunt field was wired so that it could be run in manual with a rheostat, or in auto using an automatic voltage regulator. When using the AVR, the rheostat was still in the circuit. So, the AVR was in parallel to the shunt field, and that parallel combination was in series with the rheostat.

There was a diode in the AVR which had failed, allowing the current to flow through the AVR AND the shunt field. The current measured at the rheostat (2.5A) represented the current in the shunt field and the parallel path through the AVR.